Three-dimensional (3D) video includes multiple views, each taken from a different camera perspective or angle. Each view includes a texture component (also referred to as a “texture view”) and a depth map component. 3D video encoding includes encoding the multiple texture views and their associated depth maps, and then multiplexing the multiple encoded texture views and depth maps into a bit stream. Such 3D video encoding (or compression) employs different types of video prediction, such as inter and intra prediction, to eliminate redundancy in the texture and depth map components. Such prediction is performed using a set of prediction parameters to control the prediction. Current 3D coding standards and related techniques are inflexible and restrictive with regard to the conditions under which such prediction parameters may be inherited or passed between different contexts, such as different views and their corresponding texture/depth map components, to facilitate more efficient encoding and decoding processes.
In the drawings, the leftmost digit(s) of a reference number identifies the drawing in which the reference number first appears.